System and Method for Supporting Sma Level Handling to Ensure Subnet Integrity in a High Performance Computing Environment

1. A system for supporting SMA level handling to ensure subnet integrity in a high performance computing environment, comprising: one or more microprocessors; a first subnet, the first subnet comprising one or more switches, the one or more switches comprising at least a leaf switch, wherein each of the one or more switches comprise a plurality of switch ports, a plurality of host channel adapters, each host channel adapter comprising at least one host channel adapter port, a plurality of end nodes, wherein each of the end nodes are associated with at least one host channel adapter of the plurality of host channel adapters, and a subnet manager, the subnet manager running on one of the one or more switches and the plurality of host channel adapters; wherein a switch port of the plurality of switch ports on a switch of the one or more switches is configured as a router port; wherein the switch port configured as the router port is logically connected to a virtual router; wherein the switch of the one or more switches that comprises the switch port of the plurality of switch ports configured as a router port comprises a firmware, the firmware running a subnet management agent (SMA); wherein the switch port of the plurality of switch ports configured as a router port receives, from a physical link, a subnet management packet (SMP); and wherein upon the switch port of the plurality of switch ports configured as a router port receiving the subnet management packet from a physical link, the SMA attempts to validate the SMP.

2. The system of claim 1 , wherein the SMP comprises a vendor specific SMP (VSMP); wherein the physical link comprise a first end and a second end; wherein the first end of the physical link attaches to the switch port of the plurality of switch ports configured as a router port; and wherein the second end of the physical link attaches to a second subnet.

3. The system of claim 2 , wherein the second subnet comprises: one or more switches of the second subnet, the one or more switches of the second subnet comprising at least a leaf switch of the second subnet, wherein each of the one or more switches of the second subnet comprise a plurality of switch ports of the second subnet, a plurality of host channel adapters of the second subnet, each host channel adapter of the second subnet comprising at least one host channel adapter port of the second subnet; a plurality of end nodes of the second subnet, wherein each of the end nodes of the second subnet are associated with at least one host channel adapter of the second subnet of the plurality of host channel adapters of the second subnet, and a subnet manager of the second subnet, the subnet manager of the second subnet running on one of the one or more switches of the second subnet and the plurality of host channel adapters of the second subnet; wherein a switch port of the second subnet of the plurality of switch ports of the second subnet on a switch of the another one or more switches of the second subnet is configured as a router port of the second subnet; wherein the switch port of the second subnet configured as the router port of the second subnet is logically connected to a virtual router of the second subnet, the virtual router of the second subnet comprising at least two virtual router ports of the second subnet; and wherein the second end of the physical link attaches to the switch port of the second subnet configured as the router port of the second subnet.

4. The system of claim 3 , wherein the VSMP received at the switch port configured as a router port is a direct routed (DR) VSMP, the DR VSMP comprising path information, the DR path information comprising a number of sequential hops; wherein the DR path information includes one or more hops beyond the switch port configured as a router port.

5. The system of claim 4 , wherein upon receiving the DR VSMP at the switch port configured as a router port, determines that the DR path information includes one or more hops beyond the switch port configured as a router port.

6. The system of claim 5 , wherein upon determining that the DR path information includes one or more hops beyond the switch port configured as a router port, the SMA drops the DR VSMP.

7. The system of claim 6 , wherein upon dropping the DR VSMP, the SMA sends an error message on the physical link to the second subnet; wherein the error message indicates that the DR VSMP extended one or more hops beyond the switch port configured as a router port.

8. A method for supporting SMA level handling to ensure subnet integrity in a high performance computing environment, comprising: providing, at one or more computers, including one or more microprocessors, a first subnet, the first subnet comprising one or more switches, the one or more switches comprising at least a leaf switch, wherein each of the one or more switches comprise a plurality of switch ports, a plurality of host channel adapters, each host channel adapter comprising at least one host channel adapter port, a plurality of end nodes, wherein each of the end nodes are associated with at least one host channel adapter of the plurality of host channel adapters, and a subnet manager, the subnet manager running on one of the one or more switches and the plurality of host channel adapters; configuring a switch port of the plurality of switch ports on a switch of the one or more switches as a router port; logically connecting the switch port configured as the router port to a virtual router; running a subnet management agent on a firmware of the switch of the one or more switches that comprises the switch port of the plurality of switch ports configured as a router port; receiving, at the switch port of the plurality of switch ports configured as a router port, from a physical link, a subnet management packet (SMP); and upon the switch port of the plurality of switch ports configured as a router port receiving the subnet management packet from a physical link, validating, by the SMA, the SMP.

9. The method of claim 8 , further comprising: wherein the SMP comprises a vendor specific SMP (VSMP); wherein the physical link comprise a first end and a second end; attaching the first end of the physical link to the switch port of the plurality of switch ports configured as a router port; and attaching the second end of the physical link to a second subnet.

10. The method of claim 9 , further comprising: providing, at one or more computers, including one or more microprocessors, the second subnet, the second subnet comprising: one or more switches of the second subnet, the one or more switches of the second subnet comprising at least a leaf switch of the second subnet, wherein each of the one or more switches of the second subnet comprise a plurality of switch ports of the second subnet, a plurality of host channel adapters of the second subnet, each host channel adapter of the second subnet comprising at least one host channel adapter port of the second subnet; a plurality of end nodes of the second subnet, wherein each of the end nodes of the second subnet are associated with at least one host channel adapter of the second subnet of the plurality of host channel adapters of the second subnet, and a subnet manager of the second subnet, the subnet manager of the second subnet running on one of the one or more switches of the second subnet and the plurality of host channel adapters of the second subnet; configuring a switch port of the second subnet of the plurality of switch ports of the second subnet on a switch of the another one or more switches of the second subnet as a router port of the second subnet; logically connecting the switch port of the second subnet configured as the router port of the second subnet to a virtual router of the second subnet, the virtual router of the second subnet comprising at least two virtual router ports of the second subnet; and attaching the second end of the physical link attaches to the switch port of the second subnet configured as the router port of the second subnet.

11. The method of claim 10 , the steps further comprising: wherein the VSMP received at the switch port configured as a router port is a direct routed (DR) VSMP, the DR VSMP comprising path information, the DR path information comprising a number of sequential hops; wherein the DR path information includes one or more hops beyond the switch port configured as a router port.

12. The method of claim 11 , further comprising: upon receiving the DR VSMP at the switch port configured as a router port, determining, by the SMA, that the DR path information includes one or more hops beyond the switch port configured as a router port.

13. The method of claim 12 , further comprising upon determining that the DR path information includes one or more hops beyond the switch port configured as a router port, dropping, by the SMA, the DR VSMP.

14. The method of claim 13 , further comprising: upon dropping the DR VSMP, sending, by the SMA, an error message on the physical link to the second subnet; and wherein the error message indicates that the DR VSMP extended one or more hops beyond the switch port configured as a router port.

15. A non-transitory computer readable storage medium, including instructions stored thereon for supporting SMA level handling to ensure subnet integrity in a high performance computing environment, which when read and executed by one or more computers cause the one or more computers to perform steps comprising: providing, at one or more computers, including one or more microprocessors, a first subnet, the first subnet comprising one or more switches, the one or more switches comprising at least a leaf switch, wherein each of the one or more switches comprise a plurality of switch ports, a plurality of host channel adapters, each host channel adapter comprising at least one host channel adapter port, a plurality of end nodes, wherein each of the end nodes are associated with at least one host channel adapter of the plurality of host channel adapters, and a subnet manager, the subnet manager running on one of the one or more switches and the plurality of host channel adapters; configuring a switch port of the plurality of switch ports on a switch of the one or more switches as a router port; logically connecting the switch port configured as the router port to a virtual router; running a subnet management agent on a firmware of the switch of the one or more switches that comprises the switch port of the plurality of switch ports configured as a router port; receiving, at the switch port of the plurality of switch ports configured as a router port, from a physical link, a subnet management packet (SMP); and upon the switch port of the plurality of switch ports configured as a router port receiving the subnet management packet from a physical link, validating, by the SMA, the SMP.

16. The non-transitory computer readable storage medium of claim 15 , the steps further comprising: wherein the SMP comprises a vendor specific SMP (VSMP); wherein the physical link comprise a first end and a second end; attaching the first end of the physical link to the switch port of the plurality of switch ports configured as a router port; and attaching the second end of the physical link to a second subnet.

17. The non-transitory computer readable storage medium of claim 16 , the steps further comprising: providing, at one or more computers, including one or more microprocessors, the second subnet, the second subnet comprising: one or more switches of the second subnet, the one or more switches of the second subnet comprising at least a leaf switch of the second subnet, wherein each of the one or more switches of the second subnet comprise a plurality of switch ports of the second subnet, a plurality of host channel adapters of the second subnet, each host channel adapter of the second subnet comprising at least one host channel adapter port of the second subnet; a plurality of end nodes of the second subnet, wherein each of the end nodes of the second subnet are associated with at least one host channel adapter of the second subnet of the plurality of host channel adapters of the second subnet, and a subnet manager of the second subnet, the subnet manager of the second subnet running on one of the one or more switches of the second subnet and the plurality of host channel adapters of the second subnet; configuring a switch port of the second subnet of the plurality of switch ports of the second subnet on a switch of the another one or more switches of the second subnet as a router port of the second subnet; logically connecting the switch port of the second subnet configured as the router port of the second subnet to a virtual router of the second subnet, the virtual router of the second subnet comprising at least two virtual router ports of the second subnet; and attaching the second end of the physical link attaches to the switch port of the second subnet configured as the router port of the second subnet.

18. The non-transitory computer readable storage medium of claim 17 , the steps further comprising: wherein the VSMP received at the switch port configured as a router port is a direct routed (DR) VSMP, the DR VSMP comprising path information, the DR path information comprising a number of sequential hops; and wherein the DR path information includes one or more hops beyond the switch port configured as a router port.

19. The non-transitory computer readable storage medium of claim 18 , the steps further comprising: upon receiving the DR VSMP at the switch port configured as a router port, determining, by the SMA, that the DR path information includes one or more hops beyond the switch port configured as a router port.

20. The non-transitory computer readable storage medium of claim 19 , the steps further comprising upon determining that the DR path information includes one or more hops beyond the switch port configured as a router port, dropping, by the SMA, the DR VSMP; upon dropping the DR VSMP, sending, by the SMA, an error message on the physical link to the second subnet; and wherein the error message indicates that the DR VSMP extended one or more hops beyond the switch port configured as a router port.